Application of the Fuzzy Approach for Evaluating and Selecting Relevant Objects, Features, and Their Ranges

Area of Science:

• Machine Learning
• Data Science
• Artificial Intelligence

Background:

• Feature selection is crucial for simplifying machine learning problems, reducing dimensionality, and enhancing computational speed.
• Existing methods may not always identify the most pertinent attributes or objects effectively.

Purpose of the Study:

• To propose novel algorithms for relevant attribute and object selection in datasets.
• To evaluate the effectiveness of a fuzzy approach in feature and object selection.
• To identify optimal value ranges for selected attributes and objects.

Main Methods:

• Development of new algorithms based on a fuzzy approach for attribute and object selection.
• Application and evaluation of the proposed methods on the Sonar dataset.
• Comparison of the fuzzy approach with traditional feature selection techniques.

Main Results:

• Preliminary results indicate a new approach to selecting relevant attributes, objects, and their value ranges.
• Detailed analysis on the Sonar dataset demonstrates the positive impact of the proposed fuzzy-based algorithms.
• The method shows potential in identifying a subset of truly relevant attributes, outperforming some traditional methods.

Conclusions:

• The proposed fuzzy-based algorithms offer an effective approach to relevant attribute and object selection.
• This method contributes to simplifying machine learning problems and improving computational efficiency.
• The findings suggest the superiority of the fuzzy approach in identifying genuinely relevant features compared to conventional methods.